K101644 · Inova Diagnostics, Inc. · MVM · Mar 23, 2011 · Immunology
Device Facts
Record ID
K101644
Device Name
QUANTA FLASH H-TTG IGG
Applicant
Inova Diagnostics, Inc.
Product Code
MVM · Immunology
Decision Date
Mar 23, 2011
Decision
SESE
Submission Type
Traditional
Regulation
21 CFR 866.5660
Device Class
Class 2
Indications for Use
The QUANTA Flash h-tTG IgG is a chemiluminescent immunoassay (CIA) for the semi-quantitative detection of IgG anti-human tissue transglutaminase (h-ITG) antibodies in human serum. The presence of IgG anti-h-tTG antibodies, in conjunction with clinical findings and other laboratory tests, can aid in the diagnosis of the gluten sensitive enteropathy celiac disease, particularly in patients with selective IgA deficiency. The QUANTA Flash™ h-tTG IgG Calibrators are intended for use with the QUANTA Flash™ h-tTG IgG chemiluminescent immunoassay (CIA) on the BIO-FLASH® instrument. Each calibrator establishes a point of reference for the working curve that is used to determine values in the measurement of IgG anti-h-tTG antibodies in serum. The QUANTA Flash™ h-tTG IgG Controls are intended for quality control purposes of the QUANTA Flash™ h-tTG IgG chemiluminescent immunoassay (CIA) kit run on a BIO-FLASH® instrument.
Device Story
QUANTA Flash h-tTG IgG is a chemiluminescent immunoassay (CIA) for detecting IgG anti-h-tTG antibodies in human serum. Used on the BIO-FLASH instrument; operated by laboratory professionals in clinical settings. Device utilizes calibrators to establish a working curve and controls for quality assurance. Output is a semi-quantitative measurement of IgG anti-h-tTG antibodies. Results assist clinicians in diagnosing celiac disease, particularly in patients with selective IgA deficiency, when combined with other clinical and laboratory findings.
Clinical Evidence
Clinical validation included 368 samples (48 CD, 320 non-CD). Sensitivity 50.0% (95% CI: 35.2-64.8%); Specificity 98.4% (95% CI: 96.4-99.5%). Method comparison with predicate (n=111) showed 90.2% positive agreement and 100% negative agreement. Analytical performance included precision (total CV 6.6-8.6%), linearity (R2 0.93-1.00), and LoD (1.7 CU).
Technological Characteristics
Chemiluminescent immunoassay (CIA) using paramagnetic beads coated with recombinant human tissue transglutaminase. Detection via isoluminol-conjugated monoclonal anti-human IgG. Instrument-based (BIO-FLASH™) with automated wash/incubation cycles. Signal measured as RLU. Reagents include sodium azide/chloramphenicol preservatives. Calibration uses lot-specific 4-parameter logistic curve stored in reagent barcode.
Indications for Use
Indicated for the semi-quantitative detection of IgG anti-human tissue transglutaminase (h-tTG) antibodies in human serum to aid in the diagnosis of celiac disease, particularly in patients with selective IgA deficiency. For prescription use only.
Regulatory Classification
Identification
A multiple autoantibodies immunological test system is a device that consists of the reagents used to measure by immunochemical techniques the autoantibodies (antibodies produced against the body's own tissues) in serum and other body fluids. Measurement of multiple autoantibodies aids in the diagnosis of autoimmune disorders (disease produced when the body's own tissues are injured by autoantibodies).
Submission Summary (Full Text)
{0}
1
# 510(k) SUBSTANTIAL EQUIVALENCE DETERMINATION DECISION SUMMARY
A. 510(k) Number:
k101644
B. Purpose for Submission:
New assay
C. Measurand:
Human Anti-Tissue Transglutaminase (tTG) Immunoglobulin G (IgG) antibodies
D. Type of Test:
Semi-quantitative chemiluminescent immunoassay
E. Applicant:
INOVA Diagnostics, Inc.
F. Proprietary and Established Names:
QUANTA Flash h-tTG IgG
QUANTA Flash™ h-tTG IgG Calibrators
QUANTA Flash™ h-tTG IgG Controls
G. Regulatory Information:
1. Regulation section:
21 CFR §866.5660, Multiple Autoantibodies Immunological Test System
21 CFR §862.1150, Calibrator
21 CFR §862.1660, Single (specified) analyte controls (assayed and unassayed)
2. Classification:
Class II (Assay and calibrator)
Class I (Controls)
3. Product code:
MVM, Autoantibodies, endomysial (tissue transglutaminase)
JIX, Calibrator, Multi-Analyte Mixture
JJX, Single (specified) analyte controls (assayed and unassayed)
4. Panel:
Immunology (82)
H. Intended Use:
1. Intended use(s):
QUANTA Flash™ h-tTG IgG:
The QUANTA Flash™ h-tTG IgG is a chemiluminescent immunoassay (CIA) for the semi-quantitative detection of IgG anti-human tissue transglutaminase (h-tTG) antibodies in human serum. The presence of IgG anti-h-tTG antibodies, in conjunction with clinical findings and other laboratory tests, can aid in the diagnosis of the gluten sensitive enteropathy celiac disease, particularly in celiac patients with selective IgA deficiency.
QUANTA Flash™ h-tTG IgG Calibrators:
The QUANTA Flash™ h-tTG IgG Calibrators are intended for use with the QUANTA Flash™ h-tTG IgG chemiluminescent immunoassay (CIA) on the BIO-FLASH™ instrument. Each calibrator establishes a point of reference for the working curve that is used to determine values in the measurement of IgG
{1}
anti-h-tTG antibodies in serum.
QUANTA Flash™ h-tTG IgG Controls:
The QUANTA Flash™ h-tTG IgG Controls are intended for quality control purposes of the QUANTA Flash™ h-tTG IgG chemiluminescent immunoassay (CIA) kit run on a BIO-FLASH™ instrument.
2. Indication(s) for use:
Same as Intended Use
3. Special conditions for use statement(s):
For prescription use only
4. Special instrument requirements:
BIO-FLASH™ Instrument System (k083518)
I. Device Description:
QUANTA Flash™ h-tTG IgG Kit contains one reagent pack (cartridge) with sufficient material for 100 tests. Each reagent pack contains the following sealed reagent tubes:
- Microparticle Reagent: 1 vial of recombinant human tissue transglutaminase coated magnetic particles preserved in a sugar and protein mixture.
- Assay Buffer: 1 vial of Tris-buffered saline with protein stabilizers and surfactant. Preservatives: sodium azide & chloramphenicol.
- Tracer IgG: 1 vial of isoluminol conjugated monoclonal anti-human IgG antibody in phosphate buffered saline with protein (bovine) stabilizer. Preservative: sodium azide.
- Sample Buffer: 1 vial of phosphate buffered saline with surfactant. Preservative: sodium azide.
The QUANTA Flash™ h-tTG IgG Kit also contains Resuspension Buffer: 1 vial of phosphate buffered saline, with glycerol, protein (bovine) stabilizer, and preservative (sodium azide).
QUANTA Flash™ h-tTG IgG Calibrators contains: four vials (two each of calibrator 1 and 2) containing human antibodies to h-tTG in a Tris-buffered saline solution with EDTA and sodium azide. Each vial contains sufficient material for 4 uses.
QUANTA Flash™ h-tTG IgG Controls contains: four vials (two each of Negative and Positive Control) containing human antibodies to h-tTG in a Tris-buffered saline solution with EDTA and sodium azide. Each vial contains sufficient material for 15 uses.
Additional Required Materials (available from INOVA Diagnostics, Inc.)
a. BIO-FLASH™ Instrument and Software System
b. BIO-FLASH™ System Rinse contains four 5 liter bottles of phosphate buffered saline with Tween-20 and sodium azide.
c. BIO-FLASH Triggers contains one bottle each of Trigger 1 (the catalyst) and 2 (the oxidant).
J. Substantial Equivalence Information:
1. Predicate device name(s):
BINDAZYME Human Anti-Tissue Transglutaminase IgG ELISA
2. Predicate K number(s):
k040466
3. Comparison with predicate:
{2}
| Similarities | | |
| --- | --- | --- |
| Item | Device | Predicate |
| Intended Use | The QUANTA Flash™ h-tTG IgG is a chemiluminescent immunoassay (CIA) for the semi-quantitative detection of IgG anti-human tissue transglutaminase (h-tTG) antibodies in human serum. The presence of IgG anti-h-tTG antibodies, in conjunction with clinical findings and other laboratory tests, can aid in the diagnosis of the gluten sensitive enteropathy celiac disease, particularly in celiac patients with selective IgA deficiency. | BINDAZYME™ Human IgG Anti-Tissue Transglutaminase EIA Kit is designed for the in vitro measurement of specific IgG autoantibodies against tissue transglutaminase (tTG) present in human serum, as an aid in the diagnosis of Celiac Disease. |
| Assay Type | Semi-quantitative immunoassay | Same |
| Analyte Detected | Human IgG anti-tissue transglutaminase autoantibodies | Same |
| Sample Matrix | Serum | Same |
| Antigen | Recombinant human tissue transglutaminase | Same |
| Differences | | |
| --- | --- | --- |
| Item | Device | Predicate |
| Cutoff | 20 units | 6 units |
| Assay Technology | Chemiluminescent Immunoassay (CIA) utilizing magnetic particles | Enzyme-linked Immunosorbent Assay (ELISA) |
| Conjugate | Isoluminol conjugated monoclonal anti-human IgG | Horse radish peroxidase conjugated rabbit anti-human IgG |
| Signal Detected | Luminescence (visible light) | Absorbance at 450nm |
| Calibration and unit calculation | Instrument specific working curve based off a 5 point lot specific master curve used for unit calculations; stored on the instrument for 30 days. | Five calibrators run each time the assay is run. |
# K. Standard/Guidance Document Referenced (if applicable):
Interference Testing in Clinical Chemistry; Approved Guideline (CLSI EP 7-A)
Method Comparison and Bias Estimation Using Patient Samples; Approved
Guideline (CLSI EP09-A2)
Evaluation of Precision Performance of Clinical Chemistry Devices; Approved
Guideline (CLSI EP5-A)
Evaluation of the Linearity of Quantitative Analytical Methods (CLSI EP6-P2)
Protocols for Determination of Limits of Detection and Limits of Quantitation (CLSI
{3}
EP17-A)
# L. Test Principle:
The principles of the QUANTA Flash™ h-tTG IgG assay are similar to many other solid phase indirect immunosorbent assays. In this case the solid phase is paramagnetic beads and the detecting reagent is an isoluminol-conjugated anti-human IgG monoclonal antibody. Specifically, a patient's serum is diluted with sample dilution buffer in a disposable cuvette. A small amount of this patient dilution is combined with assay buffer and h-tTG beads in a second cuvette, and mixed. This reaction cuvette is incubated for $9 \frac{1}{2}$ minutes at $37^{\circ}\mathrm{C}$ . The cuvette is then exposed to a small magnet that holds the beads in place, the liquid is aspirated, and the beads are resuspended as system rinse is added to the cuvette and the magnet is removed. This wash cycle is repeated one more time. During the third wash, no system rinse is added after the aspiration step, rather isoluminol conjugated monoclonal anti-human IgG (known as Tracer IgG) is added to the beads in the cuvette, and mixed. Again, the cuvette is incubated for $9 \frac{1}{2}$ minutes at $37^{\circ}\mathrm{C}$ . Three wash steps, as described in the first wash step above, are performed on the beads. In the fourth wash step, no liquid is added to the beads after the aspiration.
The cuvette is then placed in a light-tight luminometer and the beads are exposed to a catalyst and an oxidizing agent. These two reagents, or "Triggers", cause the isoluminol to produce a flash of visible light. The light produced from this reaction is measured as Relative Light Units (RLU) by the BIO-FLASH™ optical system. The RLU are proportional to the amount of bound isoluminol conjugate, which in turn is proportional to the amount of IgG anti-h-tTG antibodies bound to the h-tTG on the beads.
# M. Performance Characteristics (if/when applicable):
# 1. Analytical performance:
# a. Precision/Reproducibility:
Imprecision of the assay was evaluated in accordance with CLSI EP5-A2. Precision of the QUANTA Flash™ h-tTG IgG assay was evaluated by running seven patient samples, three were close to the cutoff of 20 CU. Samples were run using duplicate aliquots, twice a day, for 25 days over a 34 day period on one reagent lot, except for patient 8, which was run for 21 days over a 30 day period. The total number of replicates for each sample was 88.
| | | Within-Run | | Between-Run | | Between-Day | | Total | |
| --- | --- | --- | --- | --- | --- | --- | --- | --- | --- |
| Sample | Mean (CU) | SD | % CV | SD | % CV | SD | % CV | SD | % CV |
| 1 | 781.6 | 26.9 | 3.4% | 44.9 | 5.7% | 21.6 | 2.8% | 56.7 | 7.2% |
| 2 | 46.7 | 1.7 | 3.5% | 2.7 | 5.9% | 1.5 | 3.3% | 3.5 | 7.6% |
| 3 | 21.0 | 0.7 | 3.3% | 1.1 | 5.1% | 0.8 | 4.0% | 1.5 | 7.2% |
| 4 | 25.0 | 0.8 | 3.3% | 1.3 | 5.3% | 1.0 | 4.1% | 1.9 | 7.4% |
| 5 | 34.0 | 1.1 | 3.1% | 2.0 | 5.9% | 1.8 | 5.4% | 2.9 | 8.6% |
| 6 | 20.6 | 0.8 | 3.9% | 1.2 | 5.7% | 0.6 | 2.8% | 1.5 | 7.4% |
| 7 | 13.8 | 0.6 | 4.1% | 0.6 | 4.2% | 0.4 | 2.8% | 0.9 | 6.6% |
{4}
To investigate the imprecision of the assay at the extreme ends of the assay, calibrator materials were tested as samples in triplicate per run for five days (one run per day). The total number of replicates was 45.
| Lot | Sample | CU | With-in Run Range | Total |
| --- | --- | --- | --- | --- |
| 1 | Standard 1 | 3.8 | 1.3% – 6.1% | 4.9% |
| | Standard 5 | 2560 | 0.5% - 3.3% | 2.7% |
| | | | | |
| 2 | Standard 1 | 3.8 | 1.2% - 10.7% | 5.0% |
| | Standard 5 | 2560 | 0.1% - 6.5% | 4.1% |
| | | | | |
| 3 | Standard 1 | 3.8 | 0.9% - 10.7% | 6.5% |
| | Standard 5 | 2560 | 0 – 12.5% | 7.5% |
Lot-to- Lot variation was evaluated by testing several clinical samples with two lots of the assay. Lot RP0002 was tested 24 times over 12 days while RP0003 was tested six times on one day.
| Lot: | RP0002 | RP0003 | | | |
| --- | --- | --- | --- | --- | --- |
| Sample | Avg CU | Avg CU | Average | St Dev | % CV |
| G~9 | 19 | 13 | 16 | 3.8 | 23% |
| G~4 | 20 | 20 | 20 | 0.1 | 0% |
| G~6 | 57 | 56 | 56 | 0.9 | 2% |
| G~2 | 85 | 82 | 83 | 2.1 | 3% |
| G~8 | 95 | 98 | 96 | 1.9 | 2% |
| G~1 | 145 | 135 | 140 | 7.2 | 5% |
| G~7 | 169 | 191 | 180 | 15.4 | 9% |
# b. Linearity/assay reportable range:
Seven serum samples were selected to cover the entire range of the assay (see table below for concentrations). Dilutions were made by combining the selected sample with a known negative serum sample, as described in CLSI EP6-A. The observed values were graphed against the calculated values and linear regression was performed.
| Sample | Test Range (CU) | Slope (95% CI) | Y-intercept (95% CI) | R2 |
| --- | --- | --- | --- | --- |
| 1 | 8.5 – 286.6 | 1.02 (0.97-1.057) | 2.20 (-5.2 - 9.6) | 0.99 |
| 2 | 10.0 – 191.7 | 1.03 (0.99-1.06) | -2.63 (-12.0 - 6.8) | 0.98 |
| 3 | 54.5 – 1624.8 | 1.05 (0.99 - 1.10) | -6.40 (-52.2 - 39.4) | 0.99 |
| 4 | 6.2 – 40.3 | 1.09 (0.97-1.22) | -1.25 (-4.14 -1.63) | 0.98 |
| 5 | 5.1 – 7.8 | 1.37 (0.70 - 2.05) | -2.80 (-7.08 - 1.48) | 0.93 |
| 6 | 16.0 – 468.0 | 1.03 (1.00 - 1.07) | -12.7 (-21.0 - -3.50) | 1.00 |
| 7 | 857 – 3407.0 | 1.23 (0.98 - 1.48) | -880.6 (-1481 - -280) | 0.96 |
{5}
The reportable range of the assay is defined by the lowest and highest points on the master calibration curve. The lowest point is 3.8 CU while the top point is 2560.0 CU. A separate experiment showed that the assay does not appear to demonstrate a hook effect at very high (above the measuring range) concentrations.
c. Traceability, Stability, Expected values (controls, calibrators, or methods):
Traceability: There is no reference standard for IgG anti-human tissue transglutaminase antibodies. Calibrators and controls are assigned values based on a 20 unit cutoff between positive and negative during assay development.
Calibrators: The QUANTA Flash™ h-tTG IgG assay utilizes a predefined lot-specific Master Curve that is stored in the reagent pack barcode. The QUANTA Flash™ Calibrators are designed to produce an instrument-specific Working Curve from the parameters of the Master Curve. The two calibrator values are assigned using in-house standards and a four-parameter Master Curve. The assignment values of the two calibrators are used to create a lot-specific four-parameter logistic curve, using two stored parameters from the Master Curve and two lot-specific parameters based on the calibrator values. Calibrators showed acceptable accelerated stability for 2 weeks at 37°C, translating to at least 1 year of storage at 2-8°C. The calibrators may be stored open for a maximum of 8 hours onboard the instrument.
Controls: Controls are manufactured by diluting human serum containing high-titer IgG anti-h-tTG antibodies into buffer. A target CU value is achieved through trial dilutions on a small scale. Once a dilution is selected, the control is bulked, tested, and adjusted. The product undergoes extensive final validation testing to assign a final value.
Controls showed acceptable accelerated stability for 2 weeks at 37°C, translating to at least 1 year of storage at 2-8°C. Controls may be used up to 15 times, 1 hour per use onboard the instrument.
Reagent Pack Stability: The reagent pack can be stored, unopened, at 2-8°C for 1 year based on accelerated stability testing (2 weeks at 37°C). Opened reagent packs must be stored onboard the instrument, and are stable for 60 days. The working curve is valid for 30 days. Recalibration must be done every 30 days.
Sample Stability: Serum sample stability claims and storage recommendations in the package insert are based on CLSI H18-A3 "Procedures for the Handling and Processing of Blood Specimens; Approved Guideline — Third Edition".
d. Detection limit:
Limit of Blank (LoB) was determined according to EP17-A by testing Sample Buffer with no serum 40 times; the LoB was 1.1 CU. The Limit of Detection (LoD) was determined according EP17-A; the LoD was 1.7 CU.
e. Analytical specificity:
Interfering substances: Aliquots of a negative and two positive samples were
6
{6}
mixed with normal sera containing a known concentration of an interferent: bilirubin (9.5 mg/dL), cholesterol (342 mg/dL), triglycerides (643 mg/dL), or a serum hemolyzed by freezing so as to contain a high amount of hemoglobin (estimated as 20 g/dL). Control samples were mixed with a known negative serum. The results of the 1 part sample: 2 part interferent in normal serum dilutions are summarized below:
| | Normal | | Positive 1* | | Positive 2 | |
| --- | --- | --- | --- | --- | --- | --- |
| | CU | % Control | CU | % Control | CU | % Control |
| Control | 3.8 | --- | 13.4 | --- | 114 | --- |
| Bilirubin | 3.8 | 100 % | 16.2 | 123 % | 110 | 96 % |
| Cholesterol | 3.8 | 100 % | 15.0 | 114 % | 106 | 93 % |
| Triglycerides | 8.5 | 243 % | 19.5 | 148 % | 103 | 90 % |
| Hemolyzed | 3.8 | 100 % | 12.0 | 91 % | 66 | 58 % |
* Low positive sample (33.1 CU) became negative when diluted with normal (i.e. negative) serum + interferent
A statement indicating that lipemic, icteric, or grossly hemolyzed sera should not be used is included in the package insert.
Analytical Cross-reactivity: Samples from patients with various autoimmune diseases or infectious disease markers were tested in the QUANTA Flash™ h-tTG IgG assay. One patient with CMV, 2 patients with HSV, 1 patient with ANA, and 1 patient with H. pylori IgA were positive for IgG anti-tTG. All other serum samples were negative.
| Patient Group | N | anti-tTG IgG positive |
| --- | --- | --- |
| CMV | 12 | 1 |
| HSV | 17 | 2 |
| HCV | 15 | 0 |
| RA (CCP/RF) | 10 | 0 |
| ANA | 20 | 1 |
| Crohn's disease | 17 | 0 |
| Ulcerative colitis | 9 | 0 |
| H. pylori IgA | 15 | 1 |
| Hashimoto's Thyroiditis | 15 | 0 |
| Grave's disease | 15 | 0 |
f. Assay cut-off:
The assay cutoff was determined by testing 446 clinically characterized samples—single bleeds of patients who were clearly positive or negative for celiac disease (and were not on a gluten-free diet). These samples - 117 clinically positive and 329 clinically negative - were used to adjust the cutoff to 20 CU to optimize sensitivity and specificity at 49.6% and 97.6%, respectively, in this training set.
2. Comparison studies:
{7}
a. Method comparison with predicate device:
Samples for method comparison analysis included those samples from the clinical validation studies (CD, non-CD, and dermatitis herpetiformis patients) that were within the reportable range of the assay. These samples were tested on both the QUANTA Flash™ h-tTG IgG and on the predicate ELISA.
| | Predicate tTG IgG ELISA | | | |
| --- | --- | --- | --- | --- |
| | | Positive | Negative | Total |
| QUANTA Flash™ h-tTG IgG | Positive | 37 | 0 | 37 |
| | Negative | 4 | 70 | 74 |
| | Total | 41 | 70 | 111 |
Positive agreement (37/41) = 90.2% (95% C.I. = 76.9 - 97.3%)
Negative agreement (70/70) = 100% (95% C.I. = 94.9 - 100%)
Overall agreement (107/111) = 96.4% (95% C.I. = 91.1 - 98.6%)
b. Matrix comparison:
Not applicable; serum is the only indicated matrix for this assay.
3. Clinical studies:
a. Clinical Sensitivity and Specificity:
The clinical validation study included 199 normal blood donors, 71 non-celiac disease controls, 68 samples from an academic study (18 CD and 50 non-CD controls), 23 additional CD samples from the INOVA serum library, and 7 known CD with selective IgA deficiency. These samples were tested with the QUANTA Flash™ h-tTG IgG CIA. The results of this testing are shown below:
| | Diagnosis | | | |
| --- | --- | --- | --- | --- |
| | | CD | Not CD | Total |
| QUANTA Flash™ anti-tTG IgG | Positive | 24 | 4 | 28 |
| | Negative | 24 | 316 | 340 |
| | Total | 48 | 320 | 368 |
Sensitivity (24/48) = 50.0% (95% C.I. = 35.2 - 64.8%)
Specificity (316/320) = 98.4% (95% C.I. = 96.4 - 99.5%)
b. Other clinical supportive data:
The study above contained seven samples from Celiac Disease patients with selective IgA deficiency; six samples tested positive for the QUANTA Flash™ h-tTG IgG.
4. Clinical cut-off:
See Assay Cutoff.
5. Expected values/Reference range:
The expected value in the general population is negative. However, as the incidence of CD in the normal population is about 1%, some apparently healthy,
{8}
asymptomatic individuals may test positive for tTG antibodies.
**N. Proposed Labeling:**
The labeling is sufficient and it satisfies the requirements of 21 CFR Part 809.10.
**O. Conclusion:**
The submitted information in this premarket notification is complete and supports a substantial equivalence decision.
9
Predicate graph will load when search results are available.
Embedding visualization will load when search results are available.
PDF viewer will load when search results are available.
Loading panels...
Select an item from Submissions
Click any panel, subpart, regulation, product code, or device to see details here.
Section Matches
Results will appear here.
Product Code Matches
Results will appear here.
Special Control Matches
Results will appear here.
Loading collections...
Loading
My Alerts
You will receive email notifications based on the filters and frequency you set for each alert.
Sort by:
Create Alert
Search Filters
Agent Token
Create a read-only bearer token for Claude, ChatGPT, or other agents that can call HTTP APIs.
Copy this now. It will not be shown again.
Connected apps
Apps you authorized through browser sign-in. Disconnecting revokes their access immediately.
Learn the FDA Browser
Two short videos show you everything — or skip straight to the written tutorial if you'd rather read. You can reopen this any time from the Tutorial button in the top bar.
Part 1 — Search, results, and everyday workflows 16 min
Part 2 — Embeddings: the galaxy map 3 min
1. Search: exact and fuzzy
Type a phrase like "coronary artery calcification" into the search box. You get two kinds of results. Exact results match the literal phrase — prefix searches work ("coronary artery calcificati") but suffix searches do not. Fuzzy results match on the meaning and intent of your phrase rather than the exact words, and are sorted by relevance score. Hover over the Exact or Fuzzy badge on any row to see exactly why it matched.
Use the checkboxes above the results to narrow: SaMD keeps only software-only devices, AI / ML keeps only devices with AI.
Exact vs. fuzzy search: what's the difference?
Exact matches on the literal phrase (prefix search works, suffix does not). Fuzzy matches on the meaning and intent of the phrase rather than the exact words. Hover over the badge on any row to see why it matched.
You search "coronary artery calcification" and want only software devices with AI. What two filters do you apply?
Narrow by SaMD (software-only devices), then narrow by AI/ML (devices with AI).
2. The results table
Scroll right in the results table. The intended use is extracted for you — no need to open the PDF. The device story gives a high-level snapshot of what the device does and how it's used. The AI Performance sub-table shows each output name, acceptance criteria, observed values, and development/test dataset descriptions — the same format Innolitics uses for regulatory strategy outputs, and the fastest high-level fingerprint of an AI device. It is AI-generated but has been very reliable in practice.
Where do you find a device's intended use without opening the PDF?
Scroll right in the search results table. The intended use column is extracted for you; no need to dig into the 510(k) summary PDF.
What does the AI Performance sub-table show, and why is it useful?
Output name, acceptance criteria, observed values, development dataset description, and test dataset description. It's the same format we use for regulatory strategy output and Fast 510(k) input, and the fastest high-level fingerprint of an AI device. AI-generated but reliable in practice.
3. Judging fuzzy relevance
Fuzzy results trail off in relevance as you scroll. Use three signals to decide how far down to go: the fuzzy badge explanations, the intended use column, and whether your target output (e.g., Cobb angle) still appears in the AI Performance sub-table. Once it stops appearing, you're past the relevant zone. A top hit with a low score (~0.4) and a stretched explanation is a hint the closest predicates are far away — the project may be headed for De Novo. Note the fuzzy search is a pattern match: it doesn't handle negation ("not") well, and hardware devices can appear — filter by SaMD/AI ML to cut them.
How do you judge how far down fuzzy search results to go?
Use the relevancy signals: the fuzzy badge explanations, the intended use column, and whether the target output (e.g., Cobb angle) still appears in the AI Performance sub-table. Once it stops appearing, results are trailing off in relevancy.
4. Device detail page: chat and citations
Click a device name to open its detail page: device facts on the left, a chat window on the right. Ask something like "Describe the training data". The answer carries little citation bubbles — click one to jump to the highlighted passage in the source PDF, so you can verify every AI answer against the document. There's also a Download PDF button for sharing.
How do you verify an AI chat answer on the device detail page?
Click the citation bubbles to jump to the relevant highlight in the source document.
Reading rule for every project: how many summaries do you read in full?
At least the three most relevant 510(k) or De Novo summaries, in full. After that, use targeted chat questions to confirm your memory quickly. The tool supports this professional habit — it doesn't replace it.
5. Side-by-side comparison
Select multiple rows in the results table (aim for under ~10), then open the PDF Viewer tab. Ask one question — it goes to all selected devices in parallel, each with citations. This is the fastest way to compare and contrast devices: training data, PCCP scope, how they handled adding new scanners, and so on.
What does the side-by-side PDF viewer mode do?
Select multiple devices, open the PDF viewer tab, and ask one question (e.g., "Describe the training data"). It queries all selected devices simultaneously with citations, so you can compare and contrast quickly.
6. Collections
With rows selected, go to the Collections tab and create a labeled collection (e.g., "Cobb Angle Project"). Reload that selection any time — before a client call, pull up the collection and ask questions across all of its devices at once.
How do you save a set of selected devices for later use?
Select the rows, go to the Collections tab, and create a labeled collection (e.g., "Cobb Angle Project"). You can reload the selection anytime and carry it into the PDF viewer and other tabs that support selections.
7. Product codes and the regulations tree
Click a product code in the results to jump to it in the regulations tree — identification text, sibling product codes, and devices you can open in a PDF viewer on the right. Click a regulation number to see its identification, special controls, and related product codes. You can also search by product code or regulation number at the top of the tree. Always read the special controls if any exist for your device — it broadens your search and sharpens pre-kickoff research.
What can you do from the regulations tree view?
Browse product codes and regulation numbers, read the identification text and special controls, browse sibling product codes, open device PDFs on the right, and search by product code or regulation number at the top of the tree.
8. Chart view
Click Show Chart and segment by regulation number (or product code) to see which regulations dominate your result set. Clicking a regulation takes you into the regulations tree. Great for spotting that most matches are, say, hardware laparoscopic devices — a cue to go back and filter.
How do you see which regulations dominate a search result set?
Click "Show Chart" and segment by Regulation Number. Clicking a regulation takes you to the regulations tree.
9. The predicate graph
Open the Predicates tab for a family-tree view of predicate relationships. Click a node to trace its parents and children; selections from search carry over pre-selected. Commonly predicated devices are worth reading — a lot of people predicated them for a reason. The visual lineage is also handy on client calls, e.g. to show how a predicate family evolved and justify why your predicate still holds.
In the predicate graph, why are commonly predicated devices worth reading?
A lot of people predicated them for a reason. Clicking a node traces parents and children, and selections from search carry over pre-selected.
10. Embeddings: the galaxy map
The Embeddings tab plots every matching document in a 2-D "galaxy map" where semantically similar devices cluster together. Hover or click clusters to explore, and let AI label the clusters for you. Embeddings beat product codes for grouping: two devices can carry different product codes (LLZ vs. QIH) yet do the same thing — the embedding captures the meaning of the intended use and device story. This is also exactly how retrieval-augmented generation (RAG) works under the hood, and it makes a great visual on client calls.
Try it yourself
Head to the search page and work through a few of these AI/ML fuzzy searches to build intuition: perivascular fat on CT · aortic valve calcification opportunistic screening on noncontrast CT · breast cancer prediction on digital pathology slides · autism detection · gestational age prediction · a hearing aid that can also detect a pulse · foundation model based analysis of ECG · large language models · penetration test. Watch how the relevance scores, intended use, and AI Performance tables tell you when results stop being meaningful.